Fig. 4: Differences in feature reconstruction accuracy between DMN subsystems.
From: Neural decoding of autobiographical mental image features with a general semantic model
Differences between default mode network (DMN) subsystems were identified using two-tailed signed ranks tests to compare Core-DMN vs fronto-temporal DMN (FT-DMN), Core-DMN vs medial temporal DMN (MT-DMN) and FT-DMN vs MT-DMN in their ability to reconstruct each feature (50 participants). There were 60 signed ranks tests in total (3 DMN pairs*20 features). The 60 resultant p-values were adjusted according to false discovery rate (FDR)35. The figure displays only the comparisons yielding statistically significant differences [FDR(p) = 0.05]. Significant differences were also detected for Body and Time, but these are not displayed, because neither Body nor Time was reconstructed from any DMN subsystem with significant accuracy, according to the statistical significance thresholds used in Fig. 3A. Source data are provided as a Source Data file.
